Yong-Hyun Park

2nd-year Ph.D. student at UPenn CIS · park19@seas.upenn.edu

prof_pic.jpg

I am Yong-Hyun Park (call me Yong), a second-year Ph.D. student in CIS at UPenn, fortunate to be advised by Jiatao Gu. I am currently a part-time research intern at Roblox.

Previously, I was a research intern at Sony AI and NAVER AI. I earned my B.S. and M.S. in Physics Education from SNU, where I was fortunate to be advised by Junghyo Jo.

Research interests

I work on visual world models that can understand world dynamics and continually adapt based on test-time observations.

Previously, I studied diffusion models from several perspectives: latent space, parameter space, data space, and theory.

news

Jul 14, 2026 I started a part-time research internship at Roblox, mentored by Guande He.
Feb 06, 2026 Our paper “Concept-TRAK: Understanding how diffusion models learn concepts through concept attribution” has been accepted at ICLR 2026! (Mentor: Chieh-Hsin (Jesse) Lai)
Mar 11, 2025 Our paper “Jump Your Steps: Optimizing Sampling Schedule of Discrete Diffusion Models” has been accepted at ICLR 2025! (Mentor: Chieh-Hsin (Jesse) Lai)
Jan 22, 2025 I will be joining UPenn CIS as a Ph.D. student starting Fall 2025, advised by Prof. Jiatao Gu.

selected publications

  1. Concept-TRAK: Understanding how diffusion models learn concepts through concept attribution
    Yong-Hyun Park, Chieh-Hsin Lai, Satoshi Hayakawa, and 7 more authors
    In The Fourteenth International Conference on Learning Representations, 2026
  2. Jump Your Steps: Optimizing Sampling Schedule of Discrete Diffusion Models
    Yong-Hyun Park, Chieh-Hsin Lai, Satoshi Hayakawa, and 2 more authors
    In The Thirteenth International Conference on Learning Representations, 2025
  3. Understanding the Latent Space of Diffusion Models through the Lens of Riemannian Geometry
    Yong-Hyun Park, Mingi Kwon, Jaewoong Choi, and 2 more authors
    In Advances in Neural Information Processing Systems, 2023